Abstract
Undiagnosed diabetes is a threat to public health. This study aims to identify potential variables related to undiagnosed diabetes using Andersen’s behavioral model. Baseline data including blood test data from the China Health and Retirement Longitudinal Study (CHARLS) were adopted. First, we constructed health service related variables based on Andersen model. Second, univariate analysis and multiple logistic regression were used to analyze the relations of variables to undiagnosed diabetes. The strength of relationships was presented by odds ratios (ORs) and 95% confidence intervals (CIs). Finally, the prediction of multiple logistic regression model was assessed using the Receiver Operating Characteristic (ROC) curve and the area under the ROC curve (AUC). According to diagnosis standards, 1234 respondents had diabetes, among which 560 were undiagnosed and 674 were previously diagnosed. Further analysis showed that the following variables were significantly associated with undiagnosed diabetes: age as the predisposing factor; medical insurance, residential places and geographical regions as enabling factors; having other chronic diseases and self-perceived health status as need factors. Moreover, the prediction of regression model was assessed well in the form of ROC and AUC. Andersen model provided a theoretical framework for detecting variables of health service utilization, which may not only explain the undiagnosed reasons but also provide clues for policy-makers to balance health services among diverse social groups in China.
Highlights
Diabetes increases the risks of cardiovascular complications and premature death in the general population, which causes tremendous economic and social burden, and impacts the life quality of patients [1]
It was found that 5.82% (674/11,587) of the respondents reported having previously diagnosed diabetes, while 4.83% (560/11,587) of the respondents had diabetes undiagnosed
The results showed that old age (OR = 1.79, 95%confidence intervals (CIs): 1.16–2.77) compared to young age, medical insurance with the low reimbursement rate (OR = 1.64, 95%CI: 1.13–2.37) compared to that with the high reimbursement rate, living in rural areas (OR = 1.61, 95%CI: 1.24–2.11) compared to living in urban areas, living in the west (OR = 2.43, 95%CI: 1.78–3.30) or living in the middle (OR = 1.46, 95%CI: 1.08–1.96) compared to living in the east, were significantly related to undiagnosed diabetes with odds ratios (ORs) > 1, which increased risks to be undiagnosed (Table 2), while having other chronic diseases (OR = 0.41, 95%CI: 0.30–0.55) compared to no other chronic diseases and self-perceived health status as the poor (OR = 0.27, 95%CI: 0.19–0.38)
Summary
Diabetes increases the risks of cardiovascular complications and premature death in the general population, which causes tremendous economic and social burden, and impacts the life quality of patients [1]. Diabetes has been increasing globally, both in developing and developed countries with population aging, urbanization and lifestyle changes [2,3]. In China, the prevalence of diabetes presented a noticeable rise as well [9,10,11]. The results from a national survey [4] indicated that diabetes had already reached epidemic proportions in the general adult population, in particular, the prevalence of undiagnosed diabetes and pre-diabetes were underestimated and would become a major threat to public health in the near future if no measure was taken [12]. Studies of undiagnosed diabetes are of importance to early diagnosis and the reduction in complications of diabetes
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
